PAR test of cree vs citi vs cutter

robincnn said:

1825 with BJB holder VS 3590 on Ideal holder is slightly biased in favor of 1825. This is because the BJB holder acts as a small reflector.
https://www.rollitup.org/t/lens-and-reflector-optics-for-cob.893660/page-7#post-12229578
.

Click to expand...

noted i will redo at some point with either all ideals or grease+kapton to avoid reflectance.

JorgeGonzales said:

Plus cob spectrum vs Apogee response curve. Adjusting for that would move accuracy a little more in the right direction too.

Click to expand...

i use a li-cor which is pretty flat. ever so slightly weights the reds

https://www.rollitup.org/t/lets-talk-inexpensive-par-meters.911434/#post-12666128.

of course all of these discount everything over 700 which beats up on the 90 cri cob readings which are relatively strong from 700-750

...which is okay because radiation over 700 nm can not be used by photosynthesis.

i think as marcellus wallace would say "pretty fuckin far from OK"

SilvioBerlusconi said:

...which is okay because radiation over 700 nm can not be used by photosynthesis.

Click to expand...

Photosynthesis isn't everything. One example: http://link.springer.com/article/10.1007/s13580-016-0133-6

Plants are complicated yo.

[QUOTE="JorgeGonzales, post: 12724681, member: 929368"
Plants are complicated yo.[/QUOTE]

::vapes, ponders::

3000K has about the same amount of far red. I really like these ratios:


Not quite sunlight, but not bad. Probably good enough to prevent gross developmental problems in most plants, including the devil's weed. Maybe some UV would be good.

Anyway, that's where I'm heading I think. Over my lettuce, and everything else.

JorgeGonzales said:

3000K has about the same amount of far red. I really like these ratios:

View attachment 3718218
Not quite sunlight, but not bad. Probably good enough to prevent gross developmental problems in most plants, including the devil's weed. Maybe some UV would be good.

Anyway, that's where I'm heading I think. Over my lettuce, and everything else.

Click to expand...

I agree. I just got a 3000K 90CRI 1818 to play around with and I think I like it. I only have the one right now, and it's been added to a CXB3590 string so I won't really be able to do a side by side. It's become the centre COB of an X layout so hopefully it will add some extra far reds to my 3500K Crees, and it fills up the driver voltage nicely. win win I reckon.

1212 are the bomb for filling up drivers like that

MrTwist1 said:

I agree. I just got a 3000K 90CRI 1818 to play around with and I think I like it. I only have the one right now, and it's been added to a CXB3590 string so I won't really be able to do a side by side. It's become the centre COB of an X layout so hopefully it will add some extra far reds to my 3500K Crees, and it fills up the driver voltage nicely. win win I reckon.

Click to expand...

They are very similar:

I feel like I still want to add some deep reds at 660nm at some point, but otherwise I really like the colour balance on these high CRI offerings. I think there's a lot to be said for really being able to see and monitor your plants health when under high CRI lighting and in turn to be more in tune with what the pants need. Cheers

Awesome thread btw @BOBBY_G and thanks to all contributors for all the testing

MrTwist1 said:

I feel like I still want to add some deep reds at 660nm at some point, but otherwise I really like the colour balance on these high CRI offerings. I think there's a lot to be said for really being able to see and monitor your plants health when under high CRI lighting and in turn to be more in tune with what the pants need. Cheers

Click to expand...

Yep, that's about the only thing that might be needed, and even that is a "maybe".

Might have to try some citizen COB LED soon. I like the 1825 driven softly.

rhygin77 said:

I have a very novice question: should we compare between graphs, or should we consider each an independent test (because the setup or conditions might have changed between tests)? Since these are actual measurements (not theoretical calculations), I would assume they're very sensitive to distance and other conditions, unless that's controlled very carefully.

And thanks for posting these. It's awesome information.

Click to expand...

This is a great question, and got me thinking. @bobby g I think there might be a problem with your methodology. You don't want to let the temperature stabilize, otherwise you are measuring current and temp droop.

I believe you want a big heatsink with a fan on it, a room at the same temperature for all your tests, same distance to the cobs, and do a -pulsed- measurement.

This will allow all your tests to be compared fairly to one another. Obviously temp droop exists, but it's a separate variable that should be removed.

ok ill try this next time but its hard to keep the heatsinks constant temp at higher currents. default test setup is a pinfin with a 120mm fan directly on top of it blowing down

You could avoid different temps if you make a pulsed measurement with some cool-down-time. That is how the manufactures geht their data for TC 25°C. Like @JorgeGonzales already said - it would be the best approach for a comparsion.

Who cares what the performance is at 25C ? Does anybody run with a Tj at 25C ?

Isn't it more important to test closer to actual real world operating temps, like 50 C ?

I don't think so @PurpleBuz

You want to compare them at the same temperature, so the best way to achive this is doing a pulsed-measurement at room temp.
As already said, this is the best approach to get some data for a PPF efficiancy comparsion.

PurpleBuz said:

Who cares what the performance is at 25C ? Does anybody run with a Tj at 25C ?

Isn't it more important to test closer to actual real world operating temps, like 50 C ?

Click to expand...

It's actually not important at all when you are testing current droop. It's also room temperature, so easy to replicate in a controlled environment.

Anyway, I said it above. One variable at a time, not two. Basic science, controlling for variables.

To measure temperature droop, you do the opposite. Steady current, variable temperature, presumably in an expensive oven.

Actually this is how Tj temperature is best measured, and how cob manufacturers do it as far as I know. Use the cob itself as a thermometer, by first measuring the vF at varying temps with a pulsed current, getting your vF vs temperature curve, and going from there.

http://www.tek.com/sites/tek.com/files/media/document/resources/2681 Using Fwd Voltage2.pdf

Generic semiconductor method.